arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.01469cs.ROcs.AI
收藏

基于均流的单步视觉-语言-动作

Mean-Flow based One-Step Vision-Language-Action

Yang Chen, Xiaoguang Ma, Bin Zhao

AI总结:

本文提出基于均流的单步VLA方法,通过消除传统流匹配方法的一致性约束,显著提升生成效率,实现实时动作生成,实验表明其在机器人操作中的高效性。

摘要:

最近的流匹配基于视觉-语言-动作(VLA)框架在生成高频动作片段方面表现出显著优势,特别是在高度灵巧的机器人操作任务中。尽管取得了显著成就,但其实际应用受到生成延迟长的限制,这源于固有的迭代采样需求和架构限制。为解决这个关键瓶颈,我们提出了一种基于均流的单步VLA方法。具体来说,我们解决了动作生成过程中噪声引发的问题,从而消除了传统流匹配方法固有的一致性约束。这显著提高了生成效率,并实现了单步动作生成。现实中的机器人实验表明,所提出的基于均流的单步VLA的生成速度分别比SmolVLA和Diffusion Policy快8.7倍和83.9倍。这些结果阐明了其作为VLA基于机器人操作的高效骨干的巨大潜力。

英文摘要:

Recent advances in FlowMatching-based Vision-Language-Action (VLA) frameworks have demonstrated remarkable advantages in generating high-frequency action chunks, particularly for highly dexterous robotic manipulation tasks. Despite these notable achievements, their practical applications are constrained by prolonged generation latency, which stems from inherent iterative sampling requirements and architectural limitations. To address this critical bottleneck, we propose a Mean-Flow based One-Step VLA approach. Specifically, we resolve the noise-induced issues in the action generation process, thereby eliminating the consistency constraints inherent to conventional Flow-Matching methods. This significantly enhances generation efficiency and enables one-step action generation. Real-world robotic experiments show that the generation speed of the proposed Mean-Flow based One-Step VLA is 8.7 times and 83.9 times faster than that of SmolVLA and Diffusion Policy, respectively. These results elucidate its great potential as a high-efficiency backbone for VLA-based robotic manipulation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑